Cardinal Stritch University

Stritch Shares
Master's Theses, Capstones, and Projects

1-1-1978

Effects of reflective-impulsivity on reading

comprehension
Brooke Kerns

Follow this and additional works at: https://digitalcommons.stritch.edu/etd

Part of the Education Commons

Recommended Citation
Kerns, Brooke, "Effects of reflective-impulsivity on reading comprehension" (1978). Master's Theses, Capstones, and Projects. 831.
https://digitalcommons.stritch.edu/etd/831

This Research Paper is brought to you for free and open access by Stritch Shares. It has been accepted for inclusion in Master's Theses, Capstones, and
Projects by an authorized administrator of Stritch Shares. For more information, please contact smbagley@stritch.edu.
 THE EFFECTS OF

REFLECTIVE - IMPULSIVITY

ON READING COMPREHENSION

by

BROOKE KERNS

A RESEARCH PAPER

SUBMITTED IN PARTIAL FULFILLMENT OF THE

REQUIREMENTS FOR THE DEGREE OF

MASTER OF ARTS IN EDUCATION (READING SPECIALIST)

AT CARDINAL STRITCH COLLEGE

Milwaukee, Wisconsin

1978

~. .
This research paper has been

approved for the Graduate Committee

of Cardinal Stritch College by

 Acknowledgments

This research paper is dedicated to my

parents for providing my education, and to
my husband for his encouragement.
 TABLE OF CONTENTS

Chapter

I. INTRODUCTION 1

Problem
Hypothesis
Glossary
Limitations
Population Sample

II. REVIEW OF THE LITERATURE 8

History
Cognitive Style, Conceptual Tempo,
and Reading
Strategies and Their Modification
Socio-Economic Class Differences
Intelligence
Reliability and Stability of
Matching Familiar Figures

III. PROCEDURE 36

IV. RESULTS 38

Discussion

V. CONCLUSION • 43

Implications
Summary
Suggestions

BIBLIOGRAPHY • • • • 46
 FIGURES AND TABLES

Figure 1 4

Table 1 38

Table 2 39

 CHAPTER I

INTRODUCTION

Reflective - impulsivity is a dimension of cogni­

tion that refers to speed of response in problem solving.

This trait dimension was first introduced to the writer in

May, 1977, by Mrs. Dorothy Millard, school psychologist and

guidance counselor at Kennedy Middle School, Germantown,

Wisconsin.

Millard conducted a research study for a psychology

seminar at the University of Wisconsin-Milwaukee in 1972

entitled "The Effects of Impulsivity on Tests of Inductive

Reasoning_"l She used a test called Matching Familiar

Figures to identify children as either reflective or impul­

sive and compared performance on several tests of inductive

reasoning. Her hypothesis was that impulsive problem

solvers would make more errors on tests of inductive rea­

soning because they are prone to choose solutions without

careful evaluation. The results were inconclusive.

Millard thought that this was due to a small sample (11

impulsive and 11 reflective subjects).

IDorothy Millard, "The Effects of Impulsivity on

Tests of Inductive Reasoning." Unpublished research paper,
University of Wisconsin-Milwaukee, 1972.

1
 2

The writer developed an interest in conceptual tem­

po relating to reading ability soon after exposure to the

reflective - impulsive dimension of cognition. The idea to

compare reflective - impulsivity with reading comprehension,

like scattered seed, took root, grew, and flowered.

Problem

The purpose of this study is to test the signifi­

cance of the reflective - impulsivity dimension in reading

comprehension of subjects in the seventh grade.

Hypothesis

The impulsive subjects will make more errors than

the reflective subjects on the Comprehension subtest and

will have a lower grade level score than reflective sub­

jects on the Speed and Accuracy subtest of the Gates ­

1
MacGinitie Reading Tests.

Glossary of Terms

Conceptual Tempo: A dimension of cognitive style that

deals with decision time in problem solving situations.

Conceptual tempo is also referred to as behavioral

tempo.

Impulsivity: Defines a generalized tendency in children to

manifest short decision times and low accuracy on the

IGates - MacGinitie Reading Tests, Survey E,

Form 2, Teachers College Press, Teachers College, Columbia
University, New York, 1964.
 3

I
Matching Familiar Figures test. Impulsive subjects are
prone to choose solutions without careful evaluation in
solving problems with high response uncertainty.
Matching Familiar Figures: A test consisting of 12 items,
each containing a drawing of a familiar figure (the
standard) and six or eight facsimiles (see Figure 1).
The subject's task is to choose the facsimile which
exactly matches the standard. Two measures are obtained
from each subject: (a) the mean latency to the first
response on all 12 items, that is, the response time;
(b) the mean of the total number of errors on all 12
items. The Matching Familiar Figures test is abbrevi­
ated MFF. Some of the more recent research uses the
abbreviation MFFT. (See Figure 1.)
Reflectivity: This term refers to a generalized tendency
in subjects to manifest long decision times and high
accuracy on the MFF. The reflective subject carefully
evaluates all possible solutions, thus requiring more
time.
Response disposition: The tendency to respond either re­
flectively or impulsively across time and tasks.
Response latency: A measure of the time between the pre­
sentation of the problem and the first solution offered.
A basic assumption in research on reflective - impul-

IMatching Familiar Figures, Jerome Kagan, James

Hall, Harvard University, Cambridge, Massachusetts.
5

:",,:,--:-,-'::
 6

sivity is that response times are a faithful reflection

of decision times.

Response uncertainty: This term refers to problems with

many possible sOlutions. An example of low response

uncertainty would be a multiple choice question with

two possible solutions to choose from. High response

uncertainty would be a problem with six to eight pos­

sible solutions where it is not immediately obvious

which alternative is correct.

Limitations

A major limitation of this study was the size of

the sample. The Germantown School District gave the writer

l
permission to review Lorge - Thorndike scores only for

students she taught. Of the 130 seventh grade students in

this population, 25 met the criteria of a Lorge - Thorndike

2
score in the range of 100 - 110. Ault et al. concluded

that low error reliability of the MFF was remediated by

larger sample sizes, among other suggestions.

Another limitation is the testing date of the

Lorge - Thorndike. The most recent intelligence testing of

this sample was in the spring of 1975. This study was con­

lLorge - Thorndike Intelligence Tests, Houghton

Mifflin Publishing Co., 1900 South Batavia Avenue, Geneva,
Ill. 60134.
2
R. L. Aul t, c. Mi tchell, and D. P. IIartrnan, "Some
Methodological Problems in Reflection - Impulsivity Re­
search," C'hi'ld Deve'lopme'nt 47 (1976): 227-231.
 7

ducted in the fall of 1977. More recent intelligence

scores would have been preferred.

Population Sample

The Germantown School District covers an area of

52 square miles. Most of this is farmland. There are also

many subdivisions and one large condominium complex.

Germantown's diversity would average out to a middle income

population. There are three minority students in the

school district.
 CHAPTER II

REVIEW OF THE LITERATURE

History

The reflective - impulsive dimension of cognitive

style was postulated by Kagan et al. l during a study of

children's categorizing strategies. It was observed that

individuals demonstrate stylistic preference with respect

to the tempo with which they respond to visual discrimina­

tion problems.

Kagan described the impulsive child as one who may

act upon his initial hunch with little or no reflection,

think through a false hypothesis, and provide an answer

without critical evaluation.

Reflective children, in contrast, were found to

display a high standard of performance, persistence in

completing intellectual tasks, and few symptoms of hyper­

activity.

Kagan developed the Matching Familiar Figures test

as a means of identifying reflective and impulsive chil­

1
J. Kagan, B. L. Rosman, D. Day, J. Albert, and
w. Phillips, "Information Processing in the Child: Sig­
nificance of Analytic and Reflective Atti tudes," 'Psyc'ho­
logical Monographs, 78 (l964) (No.1, Whole No. 578).

8
 9

,.'

dren. He analyzed reflective and impulsive differences as

follows: "The child who impulsively selects the first
solution hypothesis that occurs to him is more likely to
produce an incorrect answer than one who reflects on the
differential validity of various possibilities."l
Studies on the MFF by Kagan, Rosman, Day, Albert
and Phillips, as reviewed by Kagan, revealed "Remarkable
intraindividual consistency in speed of decision time
across varied tasks and stability of decision time over
long and short periods.,,2

Thus, conceptual tempo was discovered, measured,

and defined. In the years that followed, psychiatry, psy­
chology, and education researched many aspects of this
cognitive dimension. Reflective - impulsivity was studied
as it related to quality of academic performance, reading,
socia-economic levels, intelligence, personality, learning
and mental disabilities, testing, strategies and their
modification, and sex differences. The MFF test originally
developed by Kagan was also analyzed as to its credibility.
The remainder of Chapter II is an attempt to review the
literature that pertains to conceptual tempo in a normal
educational setting.

1
Jerome Kagan, "Individual Differences in the
Resolution of Response Uncertainty, n Journ'a'l o'f Pe'rsona1ity
and Social Psychology 2 (1965):159.
2 Ibid ., p. 154.
 10

Cognitive Style, Conceptual Tempo,

a'n'd 'Re'a'ding
The identification and remediation of ineffective
reading behaviors is a continuous challenge for research­
ers, reading clinicians, reading specialists, and teachers.
According to Davey:
Too often we explain away poor achievement in read­
ing by either identifying a student's 'deficiency'
or by criticizing an instructional 'mismatch'. We
then set forth our plan or prescription to either
change the students or to change the instructional
setting. Improvement in the reading success of
students, however, requires a broader look at the
interaction between a student's cognitive-perceptual
characteristics and the educational environment.
Cognitive style research can contribute to a better
understanding of important student-environment 1
relationships which affect reading achievement.
In agreement with Davey is Cohn's suggestion that:
• reading difficulty, instead of being a cause
of other (academic) problems is, like them, only
an effect which is a result of a kind of cognitive
style that makes for less effective functioning in
dealing with the total environment, of which
schoolwork is only a part. 2
Cognitive style is described by Santostefano,
Rutledge, and Randall as:
Essentially, the concept of cognitive styles pro­
poses that when an individual's cognition is active
(not passive), selecting, sorting, organizing
information according to particular system princi­
ples which are influenced by motivational and

lBeth Davey, "Cognitive Styles and Reading Achieve­

ment," Journal of Reading 20 (November 1976):116.
2M• L. Cohn, "Field Dependence - Independence and
Reading Comprehension," Dissertation Abstracts 29 (1968):
477.
 11
l
personality factors.
Cognitive tempo is one aspect of cognitive style
that appears to be linked to several aspects of reading
behavior.
2
Kagan conducted the first study relating reading
ability to reflective - impulsivity. In this study 130

first and second graders were given visual matching prob­

lems, the Matching Familiar Figures test, and reading
recognition tests. The reading recognition tests consisted
of identification of letters and words. The examiner pro­
nounced a letter or word and the child had to choose from
five printed alternatives the correct answer. Kagan
hypothesized that children characteristically reflective
will commit fewer word recognition errors than impulsive
children. His results confirmed this prediction.
Kagan found the types of errors most characteristic
of the impulsive child were partial identity errors, suffix
errors, and meaningful and nonmeaningful substitutions.
The impulsive child also made responses in which
the first letter of the first or last syllable were in com­
man with the stimulus word. The child apparently had a
hypothesis about the word but offered an incorrect response

1
S. Santostefano, L. Rutledge, and D. Randall,
"Cognitive Styles and Reading Disability," Psychology in
the Schools 2 (1965):58.
2
Jerome Kagan, "Reflection - Impulsivity and Read­
ing Abili ty in Primary Grade Children," Ch'ild Devel'opment
36 (1965):609-628.
 .,'

12

due to insufficient reflection upon the validity of the

hypothesis. The impulsive children in the first grade also

had the highest reading error scores at the end of grade

two.

In a study of sixth graders relating conceptual

tempo and reading achievement, Johnson l found that reflec­

tive students scored significantly higher on reading

vocabulary and grade point indices than impulsive children.

Intelligence in this study was controlled so that statis­

tically significant differences would not affect the

results.
2
Lesiak found significant and positive relation­
ships between reflectivity and critical reading abilities.

Smith 3 also found in reflectives significant and

positive relationships for the skills of identifying main

ideas, recall of events, and reading for details. This

study compared three measures of cognitive style, one of

them the MFF, with five reading comprehension measures.

lB. L. Johnson, "Conceptual Tempo and the Achieve­

ment of Elementary School Boys." Unpublished doctoral
dissertation, Case Western Reserve University, Cleveland,
Ohio, 1968.

2J • F. Lesiak, "The Relationship of the Reflection­

Impulsivity Dimension and the Reading Ability of Elementary
School Children at Two Grade Levels." Unpublished doctoral
dissertation, The Ohio State University, Columbus, 1970.

3K• M. Smith, "The Influence of Cognitive Style and

Intelligence Variables in Aided Reading Comprehension."
Unpublished doctoral dissertation, University of Wisconsin,
Madison, 1973.
 13

The five areas tested in the Wiscon'sin Tes'ts 0'£' Re:ading

Skill Development, Level B - Comprehension were: identi­

fication of the main idea, sequence of events, predicting
outcomes, judging cause and effect, and reading for de­
tails. Recalling details in short paragraphs correlated
highly with the Matching Familiar Figures test. That read­
ing skill also correlated highly with the analytic index of
the Wechsler Intelligence Scale for Children and the Chil­
drenls Embedded Figures Test.
A study by Hood, Kendall, and Roettger, confirmed
1
by Hood and Kendall, found oral reading differences for
beginning readers of reflective and impulsive styles.
An index of reflection - impulsivity was administered
to first grade children. Samples of oral reading
behavior, Metropolitan Achievement Test scores, and
Stanford Readin'g 'Inventory word recognition test
scores were obtained. Reflective children made
fewer oral reading miscues than impulsive children,
and specifically fewer word substitution miscues.
Reflective children made more repetitions, regres­
sions, and proportionately more self corrections
than impulsive children. The oral reading rates of
the two groups were about the same • • • • In general
it appeared that the self corrections in oral read­
ing behavior of reflective children who are just
beginning to learn to read reflect an attempt to
obtain meaning, and that reflective children also
exhibit better reading comprehension in achievement
tests at the end of the first grade and possess more
extensive sight vocabularies in the first semester

lJ. E. Hood, and J. R. Kendall. "A Qualitative

Analysis of Oral Reading Miscues of Reflective and Impul­
sive Second Graders: A Follow-Up Study." Paper presented
at the annual meeting of the American Educational Research
Association, 1973.

"', ....
 14

1
of second grade.
Several studies have dealt with the relationship of
reading readiness and the reflective - impulsive dimension.
Ericson and Otto found reflective - impulsivity to
be a factor affecting the performance of kindergarten chi1­
dren on word recognition lists of similar and dissimilar
words. They found:
• impulsive children tended to do poorer than
reflective children on the reading tasks. It took
them longer to learn the lists (although the dif­
ference was not significant), they gave signifi­
cantly fewer correct word recognition responses
(p less than .10) and they tended to make more
false generalized responses (p less than .10).2
Ericson and Otto determined that the reflective
subjects performed better because they weigh the alterna­
tive hypotheses and select a word only after considering
individual letters.
Shapiro 3 concluded that visual discrimination
training could compensate for impulsive response style by

1
J. E. Hood, J. R. Kendall, and D. M. Roettger,
"An Analysis of Oral Reading Behavior of Reflective and
Impulsive Beginning Readers." Paper presented at the
annual meeting of the American Educational Research Asso­
ciation, 1973, Bethesda, Maryland, p. 19. ERIC Document
Reproduction Service, ED 078 376.
2L • Ericson, and w. Otto, "Effect of Intra-List
Similarity and Impulsivity - Reflectivity on Kindergarten
Children's Word Recogni tion Performance," Journ'al of
Educational Research 66 (July 1973) :469.
3 J • E. Shapiro, "Effects of Visual Discrimination
Training on Reading Readiness Test Performance of Impulsive
First Grade Boys," Journ'al of' 'E"d'uc'atio'nal 'R'e'sea'r'ch 69
(May 1976) :338-340.
 15

significantly increasing readiness test performance of

first grade boys.

These studies relate a wide variety of reading

skill proficiency to reflectiveness. Studies examining

strategies in conceptual tempo provide reasons for these

significant correlations.

Stra'tegies' and' Their Modification

Ault summarizes some of the differences between

reflective and i~pulsive children:

In general, reflective children have been found

to perform better on visual discrimination tasks,
serial recall, inductive reasoning, and reading
in the primary grades. l

More recent studies also show better performance

for reflective children on academic achievement and problem

solving~2 ability to attend, persist, and concentrate on

academic tasks~3 primary grade promotion~4 reading at the

lRuth L. Ault, "Problem Solving Strategies of Re­

flective, Impulsive, Fast-Accurate and Slow-Inaccurate
Children," Child Development 44 (1973):259-266.

2R • Haskins, and J. D. McKinney, "Relative Effects

of Response Tempo and Accuracy on Problem Solving and
Academic Achievement," Child Development 47 (8 '76):
690-696.

3James D. McKinney, "Problem Solving Strategies in

Reflective and Impulsive Children," Journa'l of Educ'a't'i'onal
Psychology 67 (1975):807-820.

4Stanley Messer, "Reflection - Impulsivity:

Stabi1i ty and School Failure," J'ourn'a'l 0'£ 'E'du'ca'tioh'a1
Psychology 61 (D '70):487-490.
 16

elementary and middle school levels;1,2,3 ability in ques­

tioning;4 and in arithmetic achievement. S
The task to discover why these differences exist

began to be studied in the latter half of the 1960s. Care­

ful examinations of the conceptual strategies used by

reflective and impulsive subjects gives an explanation to

the differences observed across a variety of tasks. Ault

proposes: "The difficulty impulsive children exhibit may

lie not in the speed of response, but rather in the

problem-solving strategy employed.,,6

Ault's study uncovered strategy development in 182

elementary age children using a game of 20 questions. She

lB. L. Johnson, "Conceptual Tempo and the Achieve­

ment of Elementary School Boys." Unpublished doctoral dis­
sertation, Case Western Reserve University, Cleveland,
Ohio, 1968.

2J. F. Lesiak, "The Relationship of the Reflection ­

Impulsivity Dimension and the Reading Ability of Elementary
School Children at T"110 Grade Levels." Unpublished doctoral
dissertation, The Ohio State University, Columbus, 1970.

3 K• M. Smith, "The Influence of Cognitive Style and

Intelligence Variables in Aided Reading Comprehension. I'
Unpublished doctoral dissertation, University of Wisconsin,
Madison, 1973.

4RUth L. Ault, "Problem Solving Strategies of Re­

flective, Impulsive, Fast-Accurate and Slow-Inaccurate
Children," Child Development 44 (1973) :259-266.

Sw. G. Cathcart, and W. Siedtke, "Reflectiveness/

Impulsiveness and Mathematics Achievement," The Arithmetic
Teacher 16 (November 1969) :563-567.

6RUth L. Ault, "Problem Solving Strategies of Re­

flective, Impulsive, Fast-Accurate and Slow-Inaccurate
Children," Child Development 44 (1973):259.
 17

found that impulsive subjects asked less mature questions

than reflective and fast-accurate subjects. Ault also
discovered that younger reflectives used strategies that
were comparable to those of older children labeled impul­
sive. This finding suggests an equivalence in cognitive
development at different age levels.
Kagan et al. I analyze the strategies affecting con­
ceptual tempo:
The reflective child considers the differential
validity of alternative answers, • • • and per­
sists longer with difficult tasks. The reflective
child wants to avoid making an error and inhibits
potentially incorrect hypotheses. The impulsive
child seems minimally concerned about mistakes and
makes his decisions quickly. Individual differ­
ences in time taken to evaluate a solution hypothe­
sis touch the problem-solving process in two places:
(a) selection of a solution hypothesis to act upon
and (b) evaluation of the quality of the solution
obtained. The chronology of a typical problern­
solving sequence passes through four phases. In
phase one, the problem is decoded (read or listened
to) and comprehended. In phase two, a 'best'
hypothesis is selected to act upon. Some evalua­
tion is appropriate at this point. In phase three,
the hypothesis is implemented, and, once again,
there should be some evaluation of the validity of
the solution arrived at. Finally, in phase four,
the solution is reported to an external agent. 2
A study by Drake 3 found strategies employed by im­
pulsive subjects to be very different from those classified

1
J. Kagan, L. Pearson, and L. Welch, "Conceptual
Impulsivity and Inductive Reasoning," Child Development 37
(1966):583-594.
2 Ibid ., pp. 583-584.
3
D. M. Drake, "Perceptual Correlates of Impulsive
and Reflective Behavior," Deve'lopmen'tal Ps'ychology 2
(1970):202-214.
 18

as reflective. Drake found that the impulsive subjects,

while taking the MFF, ignored some of the alternatives

over the course of the response time. Reflectives were

found to spend greater amounts of time comparing homologous

parts of different figures than did impulsives.

A later study by MCKinneyl identified four basic

levels of strategy development involving a transition from

scanning, testing one stimulus pattern at a time in an

orderly fashion, to focusing, which is testing a single

attribute hypothesis and extracting one bit of information

on each information trial. McKinney found a definite pro­

gression, with age, to the general strategy of focusing.

He commented, " . . . focusing isn't normally acquired

early in the elementary school period, nor is it completely

developed at age 11. . . . this optimal strategy was

adopted earlier by reflectives than by impulsives to use a

focusing approach in a variety of problem situations. n2

Studies by siegelman,3 Zelniker et al. 4 and

lJ. D. McKinney, "Problem Solving Strategies in

Reflective and Impulsive Children," Journal of Educational
Psychology 67 (D '75):807-820.

2 Ibid ., p. 818.

3E• Siegelman, "Reflective and Impulsive Observing

Behavior," Child Development 40 (1969):1213-1222.

4 T • Zelniker, W. E. Jeffrey, R. Ault, and J. Par­

son, "Analysis and Modification of Search Strategies of
Impulsive and Reflective Children on the Matching Familiar
Figures Test," Child Development 43 (1972):321-336.
 19

McCluskey and wright,l have also found that impulsive chil­

dren use less efficient strategies for scanning the stimu-

Ius array in matching to sample tasks.

It seems that the conceptual strategy used by re­

flectives in processing information on the MFF also enables

them to perform better on a variety of other tasks than

their impulsive counterparts. Researchers began asking

the question, "Can the impulsive tendency be modified to

encourage reflective behavior, thus increasing their level

of performance?" Just as there is "more than one way to

skin a cat," there is also more than one way to modify a

strategy.

Several experiments in modification by Albert,2

3 4
Reali and Hall, and Yanda and Kagan have shown that the

response latencies of impulsive children can be increased

by using a variety of techniques. Frequently, however,

IK. A. McCluskey, and J. L. Wright, "Age and Re­

flection - Impulsivity as Determinants of Selective and
Relevant Observant Behavior." Paper presented at the meet­
ing of the Society for Research in Child Development,
Philadelphia, March 1973.

2 J • A. Albert, "Modification of Impulsive Concep­

tual Style." (Doctoral dissertation, University of Illi­
nois, 1969.) Dissertation Abstracts International, 1970,
30 (7-B),3377 (University Microfilms No. 70-778).

3N• Reali, and V. Hall, "Effect of Success and

Failure on the Reflective and Impulsive Child," Develop­
mental Psychology 3 (1970) :392-402.

4Regina M. Yando, and Jerome Kagan, "The Effect of

Teacher Tempo on the Child," Child Development 39 (1968):
27-34.
 20

these treatments have not resulted in lower error rates.

The study by Yanda and Kagan was distinctive be­
cause it measured the effect of the teacher's conceptual
tempo on the students. The subjects were 80 boys and 80
girls randomly selected from the first grade classrooms of
ten impulsive and ten reflective female teachers with dif­
ferent years of experience. The children were tested with
the MFF at the beginning and end of the school year.
Results showed that the children taught by experienced
reflective teachers showed a greater increase in response
time over the course of the academic year than all other
children. The effect was more marked for boys than for
girls.
Many other studies involved training procedures
which attempted to teach impulsive children more efficient
information processing skills.
I
Research by Egeland involved training on the
strategy to use in situations in which several alternatives
are present simultaneously and it is not immediately obvi­
ous which alternative is correct. The strategy he taught
one group of subjects was: 1) look at the standard and all
the alternatives, 2) break the alternatives down into com­
ponent parts, 3) select one component part and compare it
across all alternatives, 4) check the standard to determine

lBryon Egeland, "Training Impulsive Children in the

Use of More Efficient Scanning Techniques," Child Deve'lop­
ment 45 (1974):165-171.
 21

the correct form of the component part, 5) successively

eliminate alternatives that deviate from the standard on
the particular component being studied, 6) continue to
eliminate alternatives based on an analysis of component
parts until only the correct alternative remains. A second

group was taught only to delay their response and a third

group was used as a control.
The results on the MFF immediately after the train­
ing indicated increased response time and decreased number
of errors for both group one and group two. Administration
of the MFF two months later showed that group one was able
to maintain its low level of errors while group two
increased in errors. The Gates - MacGinitie Reading Tests
were also administered at the beginning and end of the
school year. Group one showed improvement on the compre­
hension and vocabulary subtests while group two only
improved on vocabulary.
Other studies similar to Egeland's using the
approach of strategy training by Debus,l Ridberg et al.,2
and Heider,3 have been more successful in improving

IR. L. Debus, "Effects of Brief Observation of

Model Behavior on Conceptual Tempo of Impulsive Children,"
Developm'en'tal 'Psy'c'hology 2 (1970): 22-31.
2 E • N. Ridberg, R. D. Parke, and E. M. Hethering­
ton, "Modification of Impulsive and Reflective Cognitive
Styles Through Observation of Film - Mediated Models,"
Developmenta'l PSy'cho1ogy 5 (1971): 369-377.
3Eleanor Rosch Heider, "Information Processing and
the Modification of an Impulsive Conceptual Tempo,'" Child
Development 42 (October 1971):1276-1281.
 22

performance.
Ridberg et ale used peer modeling as a modifier.
Their subjects viewed a film of a nine year old boy
responding reflectively to the MFF. The results demon­
strated that impulsivity was inhibited.
The use of anxiety was also examined as a modifier.
One theory as to its effect is that the child who is made
anxious will more likely call correct the first or second
variant he happens to look at. The predicted effect of
anxiety arousal is, therefore, decreased response times
and increased errors.
1
A contrasting theory put forth by Messer states
that the subjects may be more careful after a few failures
and will scan the possible solutions more carefully trying
to do better. Decision times may be longer following
failure than following success.
A third alternative is that it may just depend on

the individual's reaction to anxiety.

Messer studied 60 third grade boys whose anxieties
were aroused experimentally by having children fail in an
intellectual task. The effect of the anxiety was assessed
through use of the MFF.
The induced anxiety resulted in longer decision
times for both impulsive and reflective children,

lstanley Messer, "The Effect of Anxiety Over Intel­

lectual Performance on Reflection - Impulsivity in Chil­
dren, II, 'C'hi'l'd' 'De've'l'o'pme'nt 41 (September 1970): 723-735.
 23

and in fewer errors for the impulsives who in­

creased in response time. This finding supported
the proposition that anxiety (or concern about
the quality of one's performance) is one ante­
cedent of a reflective disposition. 1
A study by Denney2 indicated that admonishment or
forced delay did not generate a greater number of higher
level questions. This research analyzed the strategies
used in questions asked by subjects to identify a drawing.
Denney concluded that, children must have requisite cogni­
tive ability and strategy if response times are to result
in improved performance.
3
The findings of Zelniker and Oppenheimer strongly
suggest that guiding impulsive children to note differ­
ences may enhance their ability to recognize words and deal
with visual stimuli in general.
Most researchers would probably agree that training
procedures which merely operate on response latency cannot
be expected to improve the quality of performance for im­
pulsive children who have not learned efficient strategies
for finding solutions. Training in an active search
strategy is the most effective method to modify an impul­
sive tempo.

lIbid., p. 723.
2D• R. Denney, "Reflection and Impulsivity as De­
terminants of Conceptual Strategy," Child Development 44
(1973):614-623.
3T• Zelniker, and L. Oppenheimer, "Modification of
Information Processing of Impulsive Children," Child
Development 44 (8 '73):445-450.
 24

All studies on modification of conceptual tempo

should be reviewed with the following warning by Margolis
et al. in mind.
If the impulsive child is to develop a more gen­
eralized reflective response pattern, he needs
sufficient opportunity to practice the wide
variety of tasks modeled. To expect that an in­
grained and pervasive tendency to respond impul­
sively will be considerably reduced after a few
modeling sessions is probably to underestimate
the strength of the disposition. l

Socio-Economic Class Differences

Investigations into conceptual tempo in the 1960s
led to the effect of socia-economic levels on reflection
and impulsivity.
Schwebel conducted the first of two studies into
this area in 1966. He studied this effect with the follow­
ing research:
Verbal performance of 30 middle- and lower-class
males (ages 9-12 years) was compared on four
standard tasks. Analysis of the data indicated
class difference in language usage is attributed
to both language-speech competence and to impul­
sivity - reflection in response. 2
On a free latency subtask the mean lower-class
latency was 2.5 seconds compared to 7.5 seconds for the
middle-class subjects. The lower-class subjects improved

lH. Margolis, G. Brannigan, and M. A. Poston,

"Modification of Impulsivity: Implications for Teaching,"
The Elementary School Journal 77 (January 1977):231-237.
2Andrew I. Schwebel, "Effects of Impulsivity on
Performance of Verbal Tasks in Middle- and Lower-Class
Children," American J'ournal of Orth'opsyc'h'iatry 36 (1966):
12-21.
 25

accuracy when modified by a forced latency.

Schwebel comments on his results:

Besides being handicapped by inadequately developed

vocabularies, the LC (lower-class) children also
were found to be hampered by a tendency toward
impulsivity. Apparently much that has been attri­
buted to 'just class differences' in the past can
be explained in terms of differences in this
variable. l
2
The results of a 1967 study by Kaplan and Mandel

directly oppose Schwebel. Their study involved an object

sorting task presented to six, eight, ten, and twelve year

old boys of middle- and lower-class. There were 12 middle-

class and 12 lower-class boys tested at each age level.

They found that free-latency scored higher in accuracy than

forced latency. Another finding in contrast to Schwebel

was that the delay of response condition did not differen­

tiate classes. Data indicated that at age six, the forced

delay impedes the conceptual behavior measured, while at

age eight, such behavior is facilitated by the condition.

This study differentiated from Schwebel's in that it did

not involve oral communication and broke the results dOTNn

by age. Kaplan and Mandel conclude that,

Significant class differences occur on those tasks

which require the child to communicate in verbal
conceptual terms • • • and where the task is fixed

1 Ibid ., p. 19.

2Marvin Kaplan, and Steven Mandel, "Class Differ­

ences in the Effects of Impulsivity, Goal Orientation, and
Verbal Expression on an Object Sorting Test." Paper pre­
sented to the Society for Research in Child Development,
New York, March 1967.
 26
1
and the child must arrive at the 'true' answer.
2
Bosco compared the visual information processing

speed of 180 elementary school children. A tachistoscope

presented a geometric shape and the interval between this

stimulus and matching of it was timed.

Data indicated that disadvantaged children required

more time to process visual information than did
middle-class children, but the processing speed for
the two groups tended to become more similar as
grade level increased. 3

The relationship between the reflective - impul­

sivity dimension of conceptual tempo and ability to inhibit

movement and intelligence was studied in 50 black middle­

4
class pre-school children by Harrison and Nadelman. The

MFF was administered along with Ma-ccoby' s DraT.., a Line

Slowly and Walk Slowly tests, which measure the ability to

inhibit motor movement upon request. The result was that

more reflective children were significantly able to inhibit

movement upon request than the impulsive children. The

ability to inhibit movement was correlated positively and

significantly with response latency and negatively and sig­

nificantly with errors. Girls were found to be more

lIbid., p. 495.

2J~es Bosco, "Visual Information Processing Speed

of Lower- and Middle-Class Children," Chi'l'd' 'Deve'lopme'nt 43

(1972):1418-1422.

3 Ibid ., p. 1418.

4 A• Harrison, and L. Nadelman, "Conceptual Tempo

and Inhibition of Movement in Black Pre-School Children,"
Child Development 43 (Je '72):657-668.
 27

reflective and better inhibitors than boys.

l
Schwebel and Bernstein conducted a study four
years after Schwebel's initial research into socia-economic
classes and reflective - impulsivity. They investigated
performance on subtests of the Wechsler Intell'igence S'cale
for Children by 18 lower-class boys who ranged in age from
9-14. The WISC was administered under two conditions, the
standard one and one in which an imposed latency period
was included.
These boys generally did better in the latter con­
dition, suggesting that the nature of the stimulus 2
task affected the quality of the response produced.
The authors believe that intelligence is formed by
transactions with the environment and self-attitudes re­
suIting from transactions. "Lower-class children experi­
ence frequent failure in their transactions with the school
environment. ,,3 A child with this history most likely
adapts an impulsive response. Impulsive responding occurs
because disappointment of erring is less if only a minimal
effort toward success is committed. This, in turn, inter­
feres with effective problem solving. The authors believe
in altering the administration procedure of the WIse in
order to regulate the impulsive response style.

lAndrew I. Schwebel, and Andrew J. Bernstein, "The

Effects of Impulsivity on the Performance of Lower-Class
Children on Four WISC Subtests," Ame'r'ic'an Journ-a'l of
Orthopsychiatry 40 (July 1970) :629-635.
2Ibid ., p. 629. 3 Ibid ., p. 630.

4 ~'" i
."": <
 28

'I"n't'e"l'l'ig'e'n'ce
Research by Kagan et al. l in 1964 uncovered a low
insignificant positive correlation between response latency
and IQ and a negative correlation between errors and IQ.
2
A study by Harrison and Nadelman supported the relation­

ship between errors and IQ.

Lewis et al. 3 found correlations between IQ and
latency ranging from .30 to .45, and for intelligence and
errors ranging from -.40 to -.67. Lewis et ale used
brighter subjects than Harrison and Nadelman did and an
easier form of the MFF. The differences in these variables
may be influential in the correlation with latency.
In a study of conceptual strategy Denney comments:
There were substantial correlations between the
measures of conceptual strategy and Lorge ­
Thorndike IQ scores. The measures of cognitive
tempo, on the other hand (latency .07, errors
-.14) were uniformly not correlated with IQ
scores, which suggest that some independent
factor contained within the Lorge - Thor'nd'ike
score contributes substantially to the child's

1
J. Kagan, B. L. Rosman, D. Day, J. Albert, and
w. Phillips, "Information Processing in the Child: Sig­
nificance of Analytic and Reflective Attitudes," Psycho­
logical Monographs 78 (1964) (No.1, Whole No. 578).
2A • Harrison, and L. Nadelman, "Conceptual Tempo
and Inhibition of Movement in Black Pre-School Children,"
Child Development 43 (Je '72):657-668.
3M• Lewis, M. Rausch, L. Goldberg, and C. Dodd,
"Error, Response Time and IQ: Sex Differences in Cognitive
Style of Pre-School Children," Perceptual and Mo'tor 'Sk'ills
26 (1968) :563-568.
 29

1
conceptual strategy.
He also states that cognitive tempo is not the only factor
underlying conceptual strategy.
Intelligence therefore does correlate with concep­
tual tempo and strategies.
An examination into how conceptual tempo affects

performance on intelligence tests was conducted by Schwebel

and Bernstein,2 who produced better performance on four
WISC subtests after an imposed latency. The researchers
noted:
The relationship between response latency and
quality (of performance) has been observed in
tasks requiring the same types of skills as those
demanded by items on the traditional intelligence
tests: visual recognition and analysis, reading
recognition ability, inductive reasoning, verbal
abilities, perceptual motor coordination, and
concept formation. 3
They conclude that, " • • • impulsiveness can sig­
nificantly affect performance on intelligence tests.,,4
Margolis et ale suggest implications of the adverse
effect of impulsivity:
Test responses are often the chief basis for diag­
nosis, yet many educational evaluations of children

ID. R. Denney, "Reflection and Impulsivity as De­

terminants of Conceptual Strategy," Child Devel'opment 44
(1973):614-623.
2
Andrew I. Schwebel, and Andrew J. Bernstein, "The
Effect of Impulsivity on the Performance of Lower-Class
Children on Four WISe Subtests," Ame'r'i'can Jour'n'a'l of
Orthopsychiatry 40 (July 1970):629-635.
3Ibid ., p. 630. 4 Ibid ., p. 634.
 30

who have learning difficulties fail to take into

account the length of time the child took to re­
flect on the alternative solutions available.
Consequently, remedial programs rarely stress
systematic procedures for reducing impulsivity.
• • • The result is a prescription of remedial
activities not addressed to the specific vari­
ables that impede learning. l
2
An earlier study by Margolis and Brannigan sug­
gests exploring the ramifications of establishing:
(a) local norms for the Matchinc; Fami'liar 'F'igu'res
test in order to assist in 1nitial1y identify­
ing impulsive children
(b) local predictive norms for impulsive and re­
flective children on response uncertainty
tests judged sensitive to the influences of
impulsivity
(c) local norms for particular tests based on
modified administration procedures designed
to reduce the influences of impulsivity.
Such norms may help mitigate the impulsivity
dilemma by providing more accurate assessments of abili­
ties.
A suggestion by Schwebel 3 is that "planning of cur­
ricula could include training, perhaps by a conditioning
process to think before responding."

IH. Margolis, G. Brannigan, and M. A. Poston,

"Modification of Impulsivity: Implications for Teaching,"
Elementary School Journal 77 (January 1977):231-237.
2H • Margolis, and G. Brannigan, "Conceptual Impul­
sivity as a Consideration in Test Interpretation,"
Psychology in the Schools-13 (October 1976):484-486.
3Andrew I. Schwebel, "Effects of Impulsivity on
Perfo~ance of Verbal Tasks in Middle- and Lower-Class
Children," American Journal of Orthopsychiatry 36 (1966):
12-21.
 31

Rel'i-a'b'i:l:i't'y' a'nod: 'S't'a'b'i'l'i'ty' of

Matching Familiar Figures

Reflective - impulsivity was postulated as a cog­

nitive disposition as a result of studies by Kagan et al. l

in 1964. Kagan later commented on those results: "These

studies revealed remarkable intraindividual consistency in

speed of decision time across varied tasks and stability

of decision time over long and short periods. "2 The

studies used a matching to sample task that then became

the MFF. Different forms were developed for children, pre­

school children, and adults.

A double median split procedure is used to deter­

mine conceptual tempo. Reflectives are above the median

on time and below the median on errors for the specific

sample of subjects taking the MFF. Impulsive subjects are

those whose time is below the median and whose errors are

above the median. Categories for fast-accurates and slow­

inaccurates also exist.

Kagan 3 investigated long term stability in 1965 in

I
J. Kagan, B. L. Rosman, D. Day, J. Albert, and
w. Phillips, "Information Processing in the Child: Sig­
nificance of Analytic and Reflective Attitudes," Psycho­
logical Monographs 78 (1964) (No.1, Whole No. 578).

2Jerome Kagan, "Individual Differences in the Reso­

lution of Response Uncertainty," Journal of Personality and
Social Psychology 2 (1965):154.

3Jerome Kagan, "Impulsive and Reflective Children:

Significance of Conceptual Tempo." In J. Krumboltz (ed.),
Learning and the Educational Process (Chicago: Rand­
McNally, 1965).
 32

which the MFF was administered to a group of subjects in

grade one and retested in grade two. A second group tested

in grade three and then retested in grade four. The total

groups' correlations ranged from .48 to over .60. The cor­
relations for error scores were .25 for boys and .51 for
girls.
l
Yando and Kagan studied the stability of average
response time for first graders on two versions of the MFF
resulting in correlations of .70 for girls and .13 for
boys. Error scores were low, .23 for girls and .24 for
boys. These tests were administered in the fall (Form F)
and in the spring (Form S).
Short term stability was also tested by Yando and
2
Kagan in which second grade children were shown a standard
and two variants with an additional variant added each week
for ten weeks. The median correlation across the ten weeks
was .68 for errors and .73 for response time.
3
Messer gave the MFF to first grade boys and then
administered a more difficult form to the same subjects
2-1/2 years later. His correlations were significant for
latency .31 and errors .33.

I
R. M. Yanda and J. Kagan, "The Effects of Teacher
Tempo on the Child," Child D'evelopment 39 (1968):27-34.
2 Ibid •

3S • Messer, "Reflection - Impulsivity: Stability

and School Failure," Jour'n'a'l of' 'E'du'cational Psy'cholo'gy
61 (1970) :487-490.
 33

1
Adams found reliability coefficients ranging from

.39 (errors for girls) to .58 (combined latencies for both

sexes) on a study of 40 six year olds tested three weeks
apart on the same form of the MFF.
These studies all reveal low but statistically sig­
nificant correlations for short and long term reliability
and stability. A group of studies in the mid 1970s began
criticizing the psychometric credibility of the MFF.
2
Block et al. argued that reflection and impul­
sivity are a function of accuracy and not of response
tempo. Their study attempted to correlate performance on
the MFF with psychological characteristics. Their sub­

jects were 100 children between the ages of 48 and 61

months. The MFF and the California Child Q Set were given.
This test consists of 100 widely ranging personality-
relevant terms that are ordered, using a forced choice
method, by a trained judge to express the judge's charac­
terization of the personality of a child.
3
Kagan, and Messer reply: " . . . among pre-school,

lW. Adams, "Strategy Differences Between Reflective

and Impulsive Children," Child Development 43 (1972):
1076-1080.

2 J • Block, J. Block, and D. Harrington, "Some

Misgivings about the Matching Familiar Figures Test as a
Measure of Reflection - Impulsivity," Child Deve'lopment
47 (1976): 611-632.
3
Jerome Kagan, and Stanley Messer, "A Reply to
'Some Misgivings About the Matching Familiar Figures Test
as a Measure of Reflection - Impulsivity'," Developmental
Psychology 11 {l975):244~248.
 34

in contrast to school-age, children, longer response times

1
are not always associated with more accurate performance."
2
Bentler and MCClain believe that the double median
split procedure is weak because,
Loss of discrimination is associated with assuming
that all children in a given quadrant have identical
scores; errors of classification magnify random
score differences; the dependency on sample medians
can lead to arbitrary groupings. 3
4
Ault summarized the strengths of the MFF as having
good validity, as demonstrated over a wide variety of cog­
nitive tasks. She also feels that the latency is reliable.
The weaknesses are a low error reliability and a moderate
negative correlation between latency and errors. Ault
feels that efforts to influence error scores often fail
because of its low error reliability. She recommends reme­
diation of these weaknesses by using larger sample sizes,
adequate research designs and appropriate statistical
analyses.
S
Egeland and weinberg conducted a study attempting

lIbid., p. 246.

P. M. Bentler, and J. McClain, "A Multitrait-

Multimethod Analysis of Reflection - Impulsivity," C'hild
Development 47 (1976):218-226.
3 Ibid ., p. 221.
4
R. L. Ault, C. Mitchell, and D. P. Hartmann, "Some
Methodological Problems in Reflection - Impulsivity
Research," Child Development 47 (1976) :227-231.
SByron Egeland, and Richard A. Weinberg, "The
Matching Familiar Figures Test: A Look at its Psychometric
Credibility," Child D'ev'elopment 47 (1976) :483-491.
 35

to evaluate the psychometric credibility of the MFF.

A study of the short term (one week) reliability
of different versions of the r{FFT for boys and
girls at kindergarten, second, and fifth grades
was conducted. The relations between time 1 and
time 2 latencies and errors separately for males
and females at all three grade levels, while sig­
nificant, were low. Canonical correlations and
time retest correlations based on combining raw
time and error scores into a standard score were
approximately .65-.75. The x 2 analyses indicated
that approximately 45%-70% of the subjects classi­
fied as impulsive or reflective at time 1 were
classified in the same way at time 2. Reliability
at the kindergarten level was particularly poor
and suggests that the MFFT should not be used with
children at that age. Mean differences were found
among different forms of the MFFT and across grade
level. l
They also state that:
The lack of norms for the MFFT test performance
further confounds the interpretation problem;
since a child is classified as reflective on the
basis of scores for a particular sample of chil­
dren in a given context. 2
While the data in their study show the MFF to be
low in relation to ideal psychometric standards, the
authors admit that it compares favorably with other mea­
sures of conceptual style.

lIbid., p. 483.
2 Ibid ., p. 484.
 CHAPTER I I I

PROCEDURE

Permission to conduct this research study had to be

obtained from several sources. The first step was to write

a letter to the Director of Special Services for the Ger­

mantown School District describing the writer's proposed

study and asking permission to use Lorge - Thorndike scores

to select a sample. The proposal was sent on September 10,

1977, and permission was granted several days later to use

the scores of only the students the writer teaches.

Twenty-five students were found to have a Lorge

Thorndike score in the range of 100 - 110. This criterion

was chosen so that intelligence would be constant and not

a variable.

A proposal containing a definition of reflective ­

impulsivity, the problem, hypothesis, and procedure was

sent, on September 23, 1977, to the superintendent of

schools and to the principal of Kennedy Middle School. The

principal approved the study immediately. The superinten­

dent approved it on September 30, 1977, indicating that

parental permission would be necessary.

A letter was sent home with the 25 subjects on

October 3, 1977, addressed to their parents. The letter

36
 37

briefly described reflective - impulsivity and discussed

the purpose of the study. The testing was outlined in

detail, informing the parents on the tests to be given and

the amount of time the testing would take. All of the

letters were signed and returned within a week, indicating

that permission was granted.

The Matching Familiar Figures test was administered

individually to the subjects. Testing was done during the

subjects' free time following lunch. Usually one child was

tested each day, but on several occasions two students were

tested during that time. The test took approximately 10 ­

15 minutes per subject. Testing on the MFF was conducted

between October 10 and November 10, 1977.

Reading Comprehension was tested using the Gates ­

MacGinitie Reading Tests, Form E. The Speed and Accuracy

and Comprehension subtests were given. The Speed and

Accuracy subtest is a four minute test and the Comprehen­

sion subtest is a 25 minute test. The test was adminis­

tered during the students' reading class on November 15,

1977. Two subjects were absent on that day and were tested

on November 18, 1977. Administration time was 29 minutes.

 ·CHAPTER IV

RESULTS

Two scores were obtained for each subject on the

MFF. The total number of errors for each of the twelve

problems were averaged for each subject, as was the mean

latency to the first response. The latency scores ranged

from 12.29 to 83.75 seconds for impulsive and reflective

subjects. The errors ranged from 5 to 24. Correlation of

MFF errors to latency, using the Spearman Rank formula,

was -.73.

TABLE 1

MFF MEAN ERRORS AND LATENCY FOR REFLECTIVE

AND IMPULSIVE SUBJECTS

Mean Errors Mean Latency

Reflective 11.2 44.85 seconds

Impulsive 20.1 15.25 seconds

The range of grade level scores on the Gates -

MacGinitie Reading Tests for reflective and impulsive sub­

jects on the Comprehension subtest was 3.0 to 10.0. The

range for the Accuracy grade level scores on the Speed and

Accuracy subtest was 3.7 to 11.7. The mean grade level

score for impulsive subjects was 6.9. The mean for reflec­

38
 39

tive subjects was 5.87.

The Spearman Rank formula was used to correlate

data from the MFF with data from the Gates - MacGinitie

Reading' Tests.

TABLE 2

CORRELATION OF MFF WITH GATES ­

MacGINITIE SUBTESTS

Grade Level Score Number of Errors

Speed and Accuracy Comprehension
Subtest Subtest

MFF Errors .039 -.180

MFF Latency -.012 .389

Discussion

These results are contrary to the hypothesis, which

stated: The impulsive subjects will make more errors than

the reflective subjects on the Comprehension subtest and

will have a lower grade level score than reflective sub­

jects on the Speed and Accuracy subtest of the Gates -

MacGinitie Reading Tests.

The writer observed that Comprehension subtest

scores were lower than reading class functioning levels for

seven subjects. The Speed and Accuracy subtest grade level

scores correlated more closely with the teacher's observa­

tions. The writer is also the subjects' reading teacher,

providing daily observation of the functioning level of the

subjects. Of the seven subjects scoring lower than their

demonstrated classroom reading levels on the Comprehension

 40

subtest, two were classified as impulsive and five were

reflective.

This difference in scores was 1.7 grade levels

lower for each of the two impulsive subjects. Reflective

subjects scored 1.7, 1.9, 4.5, .9, and 4.2 grade levels

lower than their demonstrated ability in reading class.

This is a total of 3.4 grade levels lower for impulsive

subjects and 13.3 grade levels lower for reflective

subjects.

An examination of the content of the school reading

program as compared to the Gates - MacGinitie Reading Tests

content provides a possible explanation for the discrep­

ancy.

The reading curriculum at Kennedy Middle School,

Germantown, Wisconsin, uses the Houghton - Mifflin Basal

Series which provides skill work, vocabulary development,

short fiction stories, and comprehension questions at three

cognitive levels. A Reading Center is also used by all

students twelve weeks out of the school year. The Power

Reading management system is used in the Reading Center and

the skills reinforced in the classroom.

The Gates - MacGinitie Speed and Accuracy subtest

consists of sentences with the last word left blank and

four multiple choice alternatives provided. This test's

administration time is four minutes. There is a great

similarity between this subtest and a favorite program of

 41

students in the Reading Center called' 'S'RA 'R'e:ad'in'g' f'or

Understanding. This similarity could account for better

test scores on the Speed and Accuracy subtest.

The Comprehension subtest consists of expository

paragraphs with two or three blanks scattered throughout.

Four multiple choice solutions are presented for each

blank. The paragraphs cover subjects such as taxation,

insurance, agriculture, history, and science. The differ­

ence in content may explain why the subjects did not score

as highly on the Comprehension subtest as on the Speed and

Accuracy subtest.

The writer also investigated reasons for the poor

test performance of the seven subjects by interviewing some

of them and closely examining the errors they made on the

reading test. The two reflective subjects with the largest

discrepancy between their score and their class work were

asked how they thought they had done on the test. Both

replied that they thought they had done satisfactorily, but

are nervous when they take tests. One of the impulsive sub­

jects seemed to mark the answer that was a short, familiar

word, even if the sentence made little sense. It appears

that this subject failed to use word attack skills. The

remainder of these seven subjects failed to use the context

clues in the paragraph to find the correct answer. Careful

reading and a search strategy to find the clue was clearly

lacking. Proof reading of the paragraphs inserting their

 42

word choices in the blanks would have indicated to these

subjects that the paragraphs did not make sense. All of

the subjects finished the Comprehension subtest with time

to spare.

A combination of nerves, carelessness, and a dif­

ferent content than that found in most of their reading

class materials may have contributed to lower Comprehension

subtest scores.

There is one other factor that may explain the non­

significance of the correlations. Almost all of the long

term studies on reflective - impulsivity have found that

children get increasingly reflective with age. Because

there are no norms with the MFF test, reflectiveness or

impulsiveness is determined by a double median split of

that particular population sample. A subject testing re­

flective with one sample, might test as impulsive in

another sample. It is conceivable that this study's sample

did not vary in range as much as similar studies using the

l{FF. Seventh grade students may have grown more reflective

and thus eliminated some of the impulsive behavior demon­

strated by younger children. The writer did not find lit­

erature on conceptual tempo that used subjects any older

than elementary school age.

The analysis of these factors may provide insight

into the discrepancy between the hypothesis and the

results.
 CHAPTER V

CONCLUSION

This study attempted to correlate reflective ­

impulsivity with reading ability. Previous research had

strongly indicated that reflective subjects make fewer

errors, and impulsive subjects make more errors, on a

variety of tasks involving response uncertainty. Knowledge

of this literature led the writer to hypothesize that the

impulsive subjects would make more errors on a multiple

choice reading test than reflective subjects.

Two instruments were chosen to measure these fac­

tors. The Matching Familiar Figures test was used to iden­

tify reflective and impulsive students. The Gates ­

MacGinitie Reading Tests (Comprehension and Speed and Accu­

racy subtests) were used to indicate grade level scores and

to analyze the number of errors each subject made.

The results of this study did not support the

hypothesis. Correlations were nonsignificant and the im­

pulsive students actually scored higher than the reflective

subjects on the Comprehension subtest.

In the process of interpreting the results the

writer made some observations of the students' performance

on both tests.

43
 44

The search strategy used by the subjects and the

improvement in accuracy as the subjects progressed through

the !{FF test was very interesting.

The analysis of the reading tests of the students

who scored much lower than their demonstrated classroom

ability indicated a lack of use of context clues. Many of

the subjects did not employ a search strategy to locate a

clue that would indicate the answer. These subjects were

also careless in taking the test as they did not reread the

paragraphs with their answers inserted in the blanks. Had

the subjects reread they would have probably realized their

mistakes. As a result of this, the writer has developed

lesson plans teaching a context clue search strategy in

multiple choice problems.

Since the completion of the study, the writer has

been tempted to administer the MFF to students functioning

in the classroom at very low reading levels and to students

at very high reading levels to see if the low ability

readers would be impulsive.

As the literature indicates, reflective - impul­

sivity is clearly a factor in any problem solving task in

which the solution is not immediately obvious. This has

particular impact on diagnostic, intelligence, and achieve­

ment testing because the results are heavily relied upon by

researchers, teachers, administrators, and the public. The

impulsive student's true ability is often judged incor­

.•.
~ ~ " "'~'-~
 45

rectly due to his impulsive response style.

The impact of conceptual tempo on middle school

and high school age students' reading tests, intelligence

tests, and achievement tests, has not been fully re­

searched. Research in these concerns would benefit under­

standing of the dynamics of cognition.

 BIBLIOGRAPHY

Adams, w. "Strategy Differences Between Reflective and

Impulsive Children." Child Development 43
(1972):1076-1080.

Albert, J. A. "Modification of Impulsive Conceptual

Style." (Doctoral dissertation, University of
Illinois, 1969.) Dissertation Abstracts Inter­
national, 1970, 30 (7-B), 3377 (University Micro­
films No. 70-778).

Ault, Ruth L. "Problem Solving Strategies of Reflective,

Impulsive, Fast-Accurate, and Slow-Inaccurate
Children." Child Development 44 (1973):259-266.

Ault, R. L.; 14itchell, C.; and Hartmann, D. P. "Some

Methodological Problems in Reflection - Impulsivity
Research." Child Development 47 (1976) :227-231.

Bentler, P. M., and McClain, J. itA Multitrait - Multi­

method Analysis of Reflection - Impulsivity."
Child Development 47 (1976):218-226.

Block, J.; Block, J.; and Harrington, D. "Some Misgivings

About the Matching Familiar Figures Test as a Mea­
sure of Reflection - Impulsivity." Child Develop­
ment 47 (1976):611-632.

Bosco, James. "Visual In.formation Processing Speed of

Lower- and Middle-Class Children." Child Develop­
ment 43 (1972) :1418-1422.

Cathcart, W. G., and Liedtke, W. "Reflectiveness/Impul­

siveness and Mathematics Achievement." The Arith­
metic Teacher 16 (November 1969):563-567.

Cohn, M. L. "Field Dependence - Independence and Reading

Comprehension." Dissertation Abstrac·ts 29 (2A)
(1968):476-477.

Davey, Beth. "Cognitive Styles and Reading Achievement."

Journal of Reading 20 (November 1976):113-120.

46

.. '.".
 47

Debus, R. L. "Effects of Brief Observation of Model Behav­

ior on Conceptual Tempo of Impulsive Children."
Deve"!opmenta'! Psychology 2 (1970): 22-31.

Denney, D. R. "Reflection and Impulsivity as Determinants

of Conceptual Strategy." Child Development 44
(1973):614-623.

Drake, D. M. "Perceptual Correlates of Impulsive and

Reflective Behavior." Developmental Psychology
2 (1970) :202-214.

Egeland, Byron. "Training Impulsive Children in the Use of

More Efficient Scanning Techniques." Child
Development 45 (1974):165-171.

Egeland, Byron, and Weinberg, Richard A. "The Matching

Familiar Figures Test: A Look at its Psychometric
Credibility." Child Development 47 (1976):483-491.

Ericson, L., and Otto, W. "Effect of Intra-List Similarity

and Impulsivity - Reflectivity on Kindergarten
Children's Word Recognition Performance." Journal
of Educational Research 66 (July 1973) :469.

Harrison, A., and Nadelman, L. "Conceptual Tempo and

Inhibition of Movement in Black Pre-School Chil­
dren." Child Development 43 (June 1972) :657-668.

Haskins, R., and McKinney, J. D. "Relative Effects of

Response Tempo and Accuracy on Problem Solving and
Academic Achievement." Child Development 47
(September 1976):690-696.

Heider, Eleanor Rosch. "Information Processing and the

Modification of an Impulsive Conceptual Tempo."
Child Development 42 (October 1971) :1276-1281.

Hood, J. E., and Kendall, J. R. "A Qualitative Analysis of

Oral Reading Miscues of Reflective and Impulsive
Second Graders: A Follow-tJp Study." Paper pre­
sented at the annual meetinq of the American Educa­
tional Research Association, 1973.

Hood, J. E.; Kendall, J. R.; and Roettger, D. M. "An

Analysis of Oral Reading Behavior of Reflective and
Impulsive Beginning Readers." Paper presented at
the annual meeting of the American Educational
Research Association, 1973, Bethesda, Maryland.
ERIC Document Reproduction Service, ED 078 376,
1973.
 48

Johnson, B. L. "Conceptual Tempo and the Achievement of

Elementary School Boys." Unpublished doctoral
dissertation, Case Western Reserve University,
Cleveland, Ohio, 1968.
Kagan, Jerome. "Impulsive and Reflective Children: Sig­
nificance of Conceptual Tempo." In J. Krurnboltz,
(ed.). Learning and the Educational Process.
Chicago: Rand - McNally, 1965.
Kagan, J. "Individual Differences in the Resolution of
Response Uncertainty." Journal of Personality and
Social Psychology 2 (1965) :154-160.
Kagan, Jerome. "Reflection - Impulsivity and Reading
Ability in Primary Grade Children." Child Develop­
ment 36 (1965) :609-628.
Kagan, Jerome, and Messer, Stanley. "A Reply to 'Some
Misgivings About the Matching Familiar Figures
Test as a Measure of Reflection - Impulsivity'."
Developmental Psychology 11 (1975):244-248.
Kagan, J.i Pearson, L.i and Welch, L. "Conceptual Impul­
sivity and Inductive Reasoning." Child Development
37 (1966) :583-594.
Kagan, J.i Rosman, B. L.i Day, D.i Albert, J.i and
Phillips, W. "Information Processing in the Child:
Significance of Analytic and Reflective Attitudes."
Psychological Monographs 78 (1964) (No.1, Whole
No. 578).
Kaplan, Marvin, and Mandel, Steven. "Class Differences in
the Effects of Impulsivity. Goal Orientation and
Verbal Expression on an Object Sorting Test."
Paper presented to Society for Research in Child
Development, New York, March 1967.
Lesiak, J. F. "The Relationship of the Reflection - Impul­
sivity Dimension and the Reading Ability of Elemen­
tary School Children at Two Grade Levels. Unpub­
lished doctoral dissertation, The Ohio State Uni­
versity, ColurrLbus, 1970.
Lewis, M.i Rausch, M Goldberg, S.; and Dodd, C. "Error
p ;

Response Time, and IQ: Sex Differences in Cogni­

tive Style of Pre-School Children." Pe·rceptual and
Motor Skills 26 (1968) :563-568.

.. ,,~.
,,~-"
 49

Margolis, H., and Brannigan, G. "Conceptual Impulsivity

as a Consideration in Test Interpretation."
Psychology in the Schools 13 (October 1976):
484-486.

Margolis, H.~ Brannigan, G.~ and Poston, M. A. "Modifica­

tion of Impulsivity: Implications for Teaching."
The Elementary School Journal 77 (January 1977):
231-237.

McCluskey, K. A., and Wright, J. L. "Age and Reflection ­

Impulsivity as Determinants of Selective and Rele­
vant Observant Behavior." Paper presented at the
meeting of the Society for Research in Child
Development, Philadelphia, March 1973.

McKinney, James D. "Problem Solving Strategies in Reflec­

tive and Impulsive Children." Journ'a'l o'f' "Educa­
tional Psychology 67 (1975) :807-820.

Messer, Stanley. "Reflection - Impulsivity: Stability and

School Failure." Journal of Educational Psychology
61 (December 1970) :487-490.

Messer, Stanley. "The Effect of Anxiety Over Intellectual

Performance on Reflection - Impulsivity in Chil­
dren." Child Development 41 (1970):723-735.

Millard, Dorothy. "The Effects of Impulsivity on Tests of

Inductive Reasoning." Unpublished research paper,
University of Wisconsin-Milwaukee, 1972.

Reali, N., and Hall, V. "Effect of Success and Failure on

the Reflective and Impulsive Child." Developmental
Psychology 3 (1970):392-402.

Ridberg, E. N.; Parke, R. D.~ and Hetherington, E. J.

"Modification of Impulsive and Reflective Cognitive
Styles Through Observation of Film - Mediated
Models." Developmental Psychology 5 (1971):
369-377.

Santostefano, S.~ Rutledge, L.; and Randall, D. "Cognitive

Styles and Reading Disability. " Psychology' 'in the
Schools 2 (1965):57-63.

Schwebel, Andrew I. "Effects of Impulsivity in Performance

of Verbal Tasks in Middle- and Lower-Class Chil­
dren." American Journal of Orthopsychiatry 36
(1966) :12-21.
 50

Schwebel, Andrew I., and Bernstein, Andrew J. "The Effect

of Impulsivity in the Performance of Lower-Class
Children on Four WISe Subtests." .Ante'ric'an Journal
of Orthopsychiatry 40 (July 1970):629-635.
Shapiro, J. E. "Effects of Visual Discrimination Training
on Reading Readiness Test Performance of Impulsive
First Grade Boys." Journal of Edu'cat'i'on'a'l Re'search
69 (May 1976):338-340.
Siegelman, E. "Reflective and Impulsive Observing Behav­
ior." Child Development 40 (1969):1213-1222.
Smith, K. M. "The Influence of Cognitive Style and Intel­
ligence Variables in Aided Reading Comprehension."
Unpublished doctoral dissertation, University of
Wisconsin, Madison, 1973.
Yando, Regina M., and Kagan, Jerome. "The Effect of
Teacher Tempo on the Child." Child Development
39 (1968) :27-34.
Zelniker, T.; Jeffrey, W. E.; Ault, R.i and Parson, J.
"Analysis and Modification of Search Strategies of
Impulsive and Reflective Children on the Matching
Familiar Figures Test." Child Development 43
(1972) :321-336.
Zelniker, T., and Oppenheimer, L. "Modification of Infor­
mation Processing of Impulsive Children." Child
Development 44 (1973) :445-450 •

• -<

!. t " .~